Actigraphy and sleep‐monitoring devices to help differentiate dementia with Lewy bodies from Alzheimer’s disease

Abstract

AbstractBackgroundDementia with Lewy bodies (DLB) is difficult to differentiate from Alzheimer’s disease (AD) in clinical settings. Actigraphy and sleep‐monitoring devices have the potential to generate data in between clinic visits to help improve the identification of DLB.ObjectiveWe will assess the feasibility of measuring activity and sleep, including day/night reversal, in participants with DLB (n=5) and AD (n=4).MethodParticipants were asked to wear actigraphy devices on their wrists for 2 weeks and sleep‐monitoring EEG devices on their foreheads for 2 nights. Actigraphy recording errors were detected by logging instances in which the majority of acceleration values approached the maximum; reliability was evaluated by logging 15‐minute actigraphy and 1‐hour sleep‐monitoring periods with at least some valid data. Actigraphy tolerability was evaluated by examining non‐wear times. Ten variables were selected to evaluate day/night reversal.ResultAveraged across all individuals, only 0.02% of the 15‐minute actigraphy recording periods had a recording error, 100% of the 15‐minute periods had at least some valid data, and only 0.15% of the periods reflected non‐wear times. For the sleep‐monitoring device, an average of 78% of each nightly recording were considered valid. Of the ten day/night reversal variables, two sleep‐monitor variables showed the strongest differences between DLB and AD: sleep efficiency (mean [SD] = 0.56 [0.19] for DLB; mean [SD] = 0.74 [0.11] for AD; see Fig. 1) and percent sleep in REM (mean [SD] = 6.83 [9.16] for DLB; mean [SD] = 18.3 [4.13] for AD; see Fig. 2). The ratio of nighttime to daytime activity as measured by actigraphy (mean [SD] = 0.37 [0.18] for DLB; mean [SD] = 0.38 [0.08] for AD) showed strong differences in the variability of measurements between AD and DLB (see Fig. 3).ConclusionThese findings suggest that actigraphy and sleep‐monitoring devices are tolerable for participants with dementia and that the devices can reliably collect valid activity and sleep data in between clinic visits. Sleep efficiency and percentage REM sleep were lower in participants with DLB than in participants with AD. Larger studies should further explore the promise of actigraphy and sleep‐monitoring devices for differentiating DLB and AD.